Skip to main content

Full text of "Early high-strength concrete for prestressing."

See other formats


Research and Development Laboratories 

of the 
Portland Cement Association 



RESEARCH DEPARTMENT 



Bulletin 91 



Early High-Strength Concrete 

for Prestressing 



BY 



PAUL KLIK.I R 






MARCH, 195 

Chicago 



Authorize*! Reprint from 

Proceedings, World Conference on Prestressed Concrete 

Sa\ Fkw UUFORNIA 



EARLY HIGH-STRENGTH CONCRETE 

FOR PRESTRESS1NG 






Paul Klieger* 



SYNOPSIS 

The continued growth of prestressed concrete de- 
pends to a large extent on its relative economy in com- 
parison with other structural materials. The cost of 
producing high -strength concrete and the resulting 
economies which can be effected by early application 
of prestress are important factors in this economic pic - 
ture. 

This report discusses the techniques which are pre- 
sently at our disposal for producing early high -strength 
concrete t Data on other properties of high -strength con- 
crete, such as elasticity, creep, drying shrinkage, bond 
to steel, and durability are presented. 



INTRODUCTION 

The use of prestressed concrete is growing rapidly. 
Continued growth is dependent to a large extent on its 
relative economy and suitability in comparison with 
other structural materials. 

Two important factors in the economics of pre- 
stressed concrete are the cost of producing early high - 
strength concrete and the resulting economies which can 
be effected by the early application of the prestress in 
either the pretensioned or post-tensioned methods. In 
the pretensioned method, early application of prestress 
permits more efficient utilization of forms and stressing 
equipment. The earlier the concrete reaches the re- 
quired strength, the earlier the prestress can be applied. 

This report deals with the means which are presently 
at our disposal for producing early high-strength con- 
crete. The choice of method for a particular situation 
will depend upon the availability of specific materials, 
mixing equipment, and compacting equipment and on 
the relative effectiveness of the method. When con- 
sidering these means for attaining early high strength. 
Important physical properties of the concrete other than 
strength must also be kept in mind. 



•Senior Research Engineer, Applied Research Section, 
Portland Cement Association, Chicago, 



FUNDAMENTAL BASIS OF STRENGTH DEVELOPMENT 

The cementing medium In concrete is produced by 
the chemical reaction between Portland cement and 
water. The inherent strength of this medium is primar- 
ily a function of the ratio of the amounts of these two 
components, normally expressed either on a weight ba- 
sis or in gallons per sack of cement. This is the fa- 
miliar "water -cement ratio". The manner in which 
this ratio influences the strength of the paste or gel, and 
hence the strength of the concrete, is illustrated in Fig. 
1. The sol id -line curve is the familiar characterization 
of Abram's water-cement ratio law for "..... plastic, 
workable mixes ,....," or as we now realize for mixes 
fully compacted (minimum entrapped air voids). Actu- 
ally, this particular example represents the relationship 
between water-cement ratio and strength for an initially 
workable mix of fixed proportions. As the water-cement 
ratio of the mix is decreased, by decreasing the amount 
of water added, the mix becomes progressively less 
workable and full compaction by hand-placing methods 
becomes impossible, as indicated by the drop in strength 
when the water-cement ratio is decreased below a point 
characteristic of each particular mix. If at this point 
mechanical compaction is used, the strengths will con- 
tinue to increase until the particular method of mechan- 
ical compaction being used no longer produces full 
compaction. 

For fully compacted concrete, therefore, the strength 

is determined primarily by the water-cement ratio of 
the paste. The proportion of the potential or ultimate 

strength actually developed in a given concrete at a 

particular age is affected by many factors such as curing 

conditions, type of cement, and others. 

INFLUENCE OF MIX PROPORTIONS 

The amount of a given aggregate that can be ac- 
commodated in a cement paste of a particular water- 
cement ratio depends upon its influence on the 
workability of the resulting mixture. The limiting 
amount is that maximum amount which can be used and 
still attain full compaction of the concrete. In other 
words, the mix proportions (meaning the relative pro- 
proportions of cement and aggregate) are of importance 
mainly insofar as they influence the workability of the 
concrete. 

Increasing the proportion of aggregate in a mix with 
a fixed water-cement ratio progressively decreases the 
consistency and workability of the concrete. There are 
a number of advantages to using the maximum amount 
of aggregate per unit of cement consistent with avail- 
able placing techniques. Some of these advantages re- 
lative to certain physical properties of the resulting 



A5-I 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



concrete will be discussed later in this report. Of main 
interest, however, is the influence of mix proportions on 
the cement requirement of the concrete. An increase in 
the amount of aggregate per unit volume in a mix of 
fixed water-cement ratio results in a reduction in both 
the amount of cement and the amount of water per unit 
volume of concrete produced and therefore a more 
economical concrete with lower shrinkage character- 
istics. Considerably more economy, in addition to other 
benefits, can be effected by utilizing a more efficient 
placing technique, such as vibration, in order to use still 
greater amounts of aggregate. Fig. 2 shows some data 
obtained recently in our laboratory using a Type 1 port- 
land cement in concretes having net water- cement 
ratios ranging from 0.29 to 0.41 by weight. The hand- 
placed concretes required considerably more paste per 
unit volume in order to provide sufficient workability to 
enable compaction with relative ease. The cement 
contents of these hand -placed mixes ranged from 6 to 12 
sacks per cubic yard. The concretes which were placed 
by external vibration, however, contained less paste per 
unit volume and though they were dry and unworkable 
by hand -placing standards, these concretes were easily 
compacted by this external vibration. The cement con- 
tents of these vibrated concretes ranged from 5 to 9£ 
sacks per cubic yard. The reductions in cement con- 
tent afforded by a change in mix proportions and a con- 
comitant change in placing technique ranged from 1 to 
2\ sacks per cubic yard. The net saving effected by 
these reductions in cement content would depend upon 
additional costs which might be incurred in mixing these 
dry concretes, transporting and vibrating, and the cost 
of more substantial forms required for vibration. 

One other aspect of mix proportions is the proportion 
of sand in the total aggregate. Progressing from a hand- 
placing technique to more and more efficient methods 
of mechanical compaction, the amount of sand required 
for good placeability can be reduced materially. This 
reduction in sand percentage permits the use of less 
water per cubic yard for the same level of workability 
and consequently a greater amount of aggregate per unit 
of cement. The vibrated mixes had lower sand percent- 
ages than the hand -placed mixes, as shown in Fig. 2. 



ment. For prestressing, the time required to develop 
the necessary strength is of particular importance. 



1 



INFLUENCE OF CURING CONDITIONS 

Since the reaction between cement and water is 
chemical, the temperature at which the reaction takes 
place would be expected to influence the rate of the 
reaction. Both the availability of water and the tem- 
perature are extremely important aspects of curing, 
particularly with regard to early high -strength develop - 



Availability of Water 

When cement and water react, an internal defi- 
ciency of water in the system may occur unless addi- 
tional curing water is supplied^). If this deficiency (by 
self-desiccation) occurs, the rate and degree of ultimate 
hydration may be reduced. Such deficiencies are more 
likely to occur in low water-cement ratio mixes for 

prestressed concrete. 

The influence of this factor on strength development 
is greater at the later ages than it is at the age of one 
day. This effect is illustrated in Figs. 3 and 4. Fig. 3 
shows data for hand -placed concretes (1 to 2 in. slump) 
where the availability of curing water during the 24 -hr. 
period in which the test cylinders were in the molds was 
increased by two means, ft) saturating the aggregate 
prior to use, and (2) ponding the top of the cylinder im- 
mediately after casting. For the 0.42 water -cement ra- 
tio concrete the increase in strength due to these tech- 
niques was nil at one day and about 750 psi at 28 days. 
For the 0.29 water-cement ratio concrete the increase 
was about 400 psi at one day and 850 psi at 28 days. 
Fig. 4 shows data for vibrated concretes (zero slump) 
made with saturated aggregate with either damp burlap 
covering or ponding during the 24 -hr. period in the 
molds. At one day, there was little influence of the 
ponding, the increases ranging from zero at 0.42 water- 
cement ratio to about 150 psi at 0.29 water-cement 
ratio. At 28 days, there was no increase at 0, 42 and 
about 1000 psi increase at 0.29 water -cement ratio. 
These increases at the low water -cement ratio become 
significant as early as three days, as can be seen in Fig. 
4, The influence of depth of section on the efficiency 
of the ponding technique has not yet been evaluated. 
Burlap in direct contact with the surface and kept satu- 
rated would approximate the ponding treatment. 

For concrete made with lightweight aggregates i ', 
tests indicate that ponding was of no benefit. Light- 
weight aggregates absorb considerable water during 
mixing which apparently can transfer to the paste dur- 
ing hydration. 



Temperature 

Since the reaction between water and cement is a 
chemical reaction, increasing the temperature during 
curing might be expected to increase the rate of strength 
gain. To illustrate the influence of temperature on 



A5-2 



EARLY HIGH-STRENGTH CONCRETE FOR PRESTRESSING 



strength development, Fig. 5 shows the one -day strengths 
of concretes mixed and cured at different tempera- 
tures^ as percentages of the 73°F strengths. These 
concretes were cast in uninsulated steel molds and were 
stored at the temperatures indicated. Temperatures be- 
low 73°F resulted in lower one -day strengths, and tem- 
peratures above 73° F resulted in higher one -day 
strengths, than that developed at 73°F. Although later - 
age strengths are somewhat reduced by initial curing at 
elevated temperatures, no retrogression in strength oc- 
curs with age ( \ 

Acceleration of strength gain at the early ages by 
providing elevated temperatures can be accomplished 
in a number of ways. The use of saturated steam at at- 
mospheric pressure serves up to the boiling point of 
water. From that point, saturated steam under pressure 
can be used to provide still higher temperatures. For 
the prestressed concrete industry, saturated steam at at- 
mospheric pressure appears most adaptable for the size f 
thape, and manner in which the majority of units must 
be made. Ponding of exposed top surfaces or covering 
with burlap kept saturated could be employed to ad- 
vantage to reduce the possibility of drying and to pro- 
vide additional curing water to counter the effect of 
lelf-dessication. Saturated steam must be provided to 
prevent drying out during the curing process. This type 
of curing normally employs temperatures in the range of 
140 to 165° F. Temperatures higher than this may, in 
some cases, result in somewhat lower strengths at the 
end of the curing cycle. This range appears to be the 
optimum for most concretes. The cycle usually con- 
sists of a 3 to 6 -hour preset, or waiting, period im- 
mediately after fabrication at normal temperature, fol- 
lowed by about a 16 -hour period of heating with satu- 
rated steam at the temperature within the range selected 
as optimum for the materials being used, and subsequent 
gradual cooling over about a 3 -hour period to avoid ex- 
cessive drying. The optimum temperature is .dependent 
to some extent on mix proportions, type of cement, and 
type of aggregate. In a recent series of tests, concretes 
were prepared at 0.4 and 0.5 water-cement ratios by 
weight, cement contents approximately 9^ and 7^ sacks 
of Type I Portland cement, respectively, and subjected 
to saturated steam at 140°F for 16 hours after a 4 -hour 
preset period at 73°F. The one -day strength of these 
concretes was 4080 and 2940 psi, respectively, in con- 
trast to the respective 73°F one -day strengths of 1590 
and 950 psi. This comparison demonstrates the strength 
increases that can be obtained by the use of saturated - 
steam curing at atmospheric pressure. 

The reaction between cement and water liberates 
heat. In the production of prestressed concrete elements 
mixes of low water-cement ratio and high cement con- 



tent are necessarily employed. Because ofthe high con- 
centration of cement in these mixes, a considerable 
amount of heat is liberated. This heat, if retained , 
can raise the concrete temperature by a substantial 
amount and thereby increase early strength develop- 
ment. Figure 6 shows the increase in concrete tempera- 
ture normally to be expected if the heat generated were 
fully retained. These theoretical calculations were 
based on heat-of-hydration studies of neat-cement pastes 
using the conduction-calorimeter technique ( 4 ) and are 
for a A\ gal. per sk. mix containing 8i sk. of Type I or 
Type III Portland cement per cu. yd. For example, 
using a typical Type I cement in this concrete mix at 
70° F, and retaining all of the heat generated, a rise of 
almost 60° F in concrete temperature to about 130° F at 
24 hours should be noted. Actually, the temperature 
would rise somewhat above 130° F f since the additional 
heat increases the rate of hydration and more and more 
heat would be generated as the concrete temperature 
rose, as seen by the 105°F data for both types of cement. 
Complete retention of the heat generated within the 
first 24 hours could raise the temperature ofthe concrete 
well over what might be considered optimum. However, 
almost perfect insulation would be extremely costly and 
is not likely to be used. Insulation of the blanket or bat 
type, such as spun glass, rock wool, balsam wool, and 
others, is reusable and can retain sufficient heat to pro- 
vide significant increases in the early strengths. Insu- 
lation used in combination with some artificial heat at 
the start to promote heat generation appears to offer 
possible economies in the production of prestressed 

units. 

The effectiveness of various insulating materials is 
being evaluated in a current series of tests in our labo- 
ratory. As an example, concrete cylinders were fabri- 
cated by external vibration in steel molds, using a dry 
mix at a water-cement ratio of 0.36 by weight and both 
Type I and Type HI Portland cements. The tops of the 
cylinders were sealed with a thin sheet of plastic to pre- 
vent loss of moisture. The group of molds was then 
completely wrapped in a one -inch-thick blanket of spun 
glass fibers. After one day, some of the cylinders were 
removed for immediate testing or for 2 days of ad- 
ditional curing in the moistroom at 73 F, The re- 
mainder were kept wrapped for an additional two days. 
The results of these strength tests are shown in Fig. 7. 
The one -day strength was increased from 2200 psi to 2700 
psi for the Type 1 cement, an increase of about 25%. For 
the Type III cement, the increase was about 28%. The 
temperature of the concrete at the end of the first day 
was about 90° F for the Type I cement and about 100 F 
for the Type III cement. The initial temperature of 
both concretes was about 75°F. Actually, for complete 



A5-3 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



retention of heat, these temperatures at one day should 
have been somewhat higher. Under these test con- 
ditions, this particular mode of insulation retained about 
40% of the heat generated. Nevertheless, the increase 
in the one -day strength was 23°}o for the Type I cement 
and 28% for the Type III cement --economically sig- 
nificant increases considering that the insulation can be 
reused a number of times. Maintaining insulation for 3 
days resulted in further increases in strength. The cost 
of more efficient insulation, which would retain a 
greater proportion of the heat generated, would have to 
be balanced against the cost of supplying additional 
heat by means of saturated steam or other means. A 
combination of these two techniques appears to be the 
most desirable course, particularly since some insulating 
enclosure must be provided even when additional heat Is 
supplied. More efficient, reusable insulation would re- 
sult in savings in the amounts of additional heat which 
would need to be provided. 



INFLUENCE OF TYPE OF CEMENT 



The rate of strength development of concrete can be 
varied by using different types of Portland cement. Of 
the different types in general use (ASTM Types I, n, 
III, and their counterpart air -entraining cements), ASTM 
Type III (high -early -strength portland cement) develops 
strength more rapidly than the other types. Concrete 
made with this type of cement gains strength more rap- 
idly at early ages than concrete made with Type I or 
Type II cement. However, under continuous moist cur- 
ing the strengths are about the same at ages later than 
about 3 months. Figure 8 shows some typical strength 
data of concretes made with Type I and Type III portland 
cements. The net water -cement ratio of these concretes 
was 0.35 by weight, the cement content 6i sk. per cu. 
yd. These were zero -slump concretes compacted by ex- 
ternal vibration. Curing was at 73°F t in steel molds for 
the first day and thereafter in the moistroom. The con- 
cretes made with the Type III cement were considerably 
stronger than those made with the Type I cement; the 
strength ratios of Type 111 to Type I were 1.86, 1.50, 
and 1.23 for 1, 3, and 7 days, respectively. The use of 
Type III cement coupled with the blanket insulation pre- 
viously described and saturated-steam curing at elevated 
temperature would produce concretes having a one -day 
strength well in excess of 5000 psi. 

THE INFLUENCE OF ACCELERATORS 

The hydration of cement can be accelerated by the 
addition of relatively small amounts of inorganic mater- 



ials. Of these materials, calcium chloride is the most 
commonly used accelerator. 

Calcium chloride in amounts ranging from one to 
two percent by weight of the cement provides significant 
increases in early strength development In common 
with other means of accelerating strengtn gain at the 
early ages, at28days and later the strengths of concretes 
with and without calcium chloride are approximately 
equal. Figure 9 shows the effect of V]o of calcium chlor- 
ide on the strength of a 0.35 water-cement ratio mix 
made with Type I portland cement. There is a consider- 
able increase in the one -day strengths, these increases 
decreasing with age, as indicated by the ratios at the 
ages of one, three, and seven days. 

However, a serious problem has arisen with regard 
to the use of calcium chloride in prestressed concrete. 

The presence of chloride ions in the concrete may pro- 
duce conditions favorable for corrosion of the steel. Lo- 
calized and severe corrosion under these conditions may 
occur in the steel, particularly at voids in the interface 
between the steel and the concrete, causing steel failure. 
A comprehensive study of this corrosion problem is now 
under way in our laboratories. Available evidence indi- 
cates only that when chloride ions are present, localized 
corrosion and consequent failure of steel under stress may 
take place. At present, the use of calcium chloride for 
prestressed concrete is not recommended, particularly 
for pretensioned prestressed applications. In post -ten- 
sioned work, either grouting of the steel after tension- 
ing or the presence of a bond -breaking shield might in- 
troduce a sufficient barrier to chloride ions. This possi- 
bility is being investigated. 



INFLUENCE OF COMPACTION 

It was stated earlier in this report that the strength of 
the concrete was controlled primarily by the water-cem- 
ent ratio, provided the concrete was fully compacted. 
Actually, strength is related to the voids-cement ratio, 
the voids being the sum of the volumes of water and air 
in the concrete, the volume of air being an index of the 
efficiency of compaction. The air in this case is not the 
air which we intentionally entrain in the paste by means 
of air -entraining cements or air -entraining admixtures, 
but is entrapped air resulting from the use of dry, harsh 
mixes. 

Any technique of compaction is suitable, provided 
full compaction can be attained. If full compaction is 
attained, the water-cement ratio will then be as good a 
criterion of strength as voids-cement ratio, since the 
amount of entrapped air voids would be small and exert 
little influence on the magnitude of the voids-cement 
ratio. Figure 10 shows the one-day and three -day strengths 



A5~4 






EARLY HIGH-STRENGTH CONCRETE FOR PRESTRESSING 



of concretes of four different water -cement ratios! for 
both hand -placed and vibrated concretes. Essentially 

similar strengths were produced by the two placing tech- 
niques. However, the hand -placed concretes had slumps 
in the range of one to two inches and were easily com- 
pacted by rodding. In a recent series of tests of zero- 
slump concrete by both hand-placing and vibration, the 
strength ratios of vibrated to hand -placed concrete ranged 
from 1.00 to 1.09 for about a 5% average increase in 
strength of the vibrated concretes. This indicates that 
with the dry, harsh mixes, hand-roddingof the specimens 
did not achieve complete compaction, even though ex- 
treme care was taken. 

In stiff, low water-cement ratio mixes, mechanical 
compaction must be resorted to in order to compact these 
mixes efficiently and economically. Figure 2 showed 
the economy which could be effected by the use of vi- 
bration in compacting these concretes. Three general 
types of vibration may be employed: internal, form, or 
surface vibration. Internal and form vibration are more 
generally applicable to the production of prestressed con- 
crete units. Of these two methods, internal vibration is 
to be preferred since the power is transmitted directly to 
the concrete and the portability of the equipment permits 
more flexible use during vibration. 

Opinions vary as to the influence of frequency of vi- 
bration and amplitude of vibration on strength. A recent 
series of tests were made in our laboratory using a vi- 
brating table to supply external vibration during casting 
of test cylinders. Aside from the time of vibration nec- 
essary, there appeared to be little choice as to frequen- 
cy or amplitude provided that the concrete was fully 
compacted. Some typical test results are shown in Fig. 
11. Frequencies ranging from 3600 to 11,000 vibrations 
per minute used in combination with different amplitude 
settings showed no consistent influence on the strengths 
of these zero -slump concretes. However, frequency and 
amplitude influenced considerably the duration of vibra- 
tion necessary to achieve adequate compaction. At a 
particular frequency, a larger amplitude of vibration re- 
duced the required time and, at a given amplitude, a 
higher frequency reduced the time for compaction. Ob- 
viously, from the standpoint of economy, the particular 
combination of frequency and amplitude which achieves 
full compaction in the shortest time would be desirable. 
To take advantage of this situation would necessitate 
cottly variable frequency, variable amplitude vibrating 
equipment. Equipment generally available, at least for 
internal vibration, is of the fixed -frequency type, with 
little opportunity for amplitude change. The choice of 
vibration equipment for a particular application cannot 
readily be determined by theoretical analyses. The in- 



fluence of factors such as frequency, amplitude, mix 
characteristics, and size and shape of member is diffi- 
cult to evaluate. The type of equipment required is gen- 
erally determined by trial. 

A means for producing low water-cement ratio con- 
crete is available in the patented vacuum-treatment pro- 
cess. This method of extracting water from plastic mix- 
es is limited to the use of relatively thin sections. 



INFLUENCE OF AGGREGATE CHARACTERISTICS 

Since aggregate comprises a large fraction of the 
volume of concrete, the characteristics of the aggregate 
significantly influence the properties of the concrete. Of 
immediate interest is the influence of these character- 
istics on strength. 

Grading and maximum size of aggregate affect 
strength only indirectly by their influence on the water 
requirements for a particular level of consistency. For 
the same water-cement ratios, different gradations of the 
same aggregate will produce essentially the same 
strengths. A grading which requires more water for a 
particular level of workability will, however, result nec- 
essarily in a higher cement content in a mix of fixed 
water-cement ratio. A decrease in the maximum size 
of aggregate also results in higher water requirement with 
its consequent influence on cement content. Generally, 
once a mix is selected, control is governed by the con- 
sistency of the concrete produced. Variations which oc- 
cur in grading would then be reflected in different a- 
mounts of water necessary to attain the desired consis- 
tency. This would result in changes in water -cement 
ratio and concomitant changes in strength. These vari- 
ations in strength, if not reduced by closer control of 
uniformity of grading, would require a redesign of the 
mix in order to insure the production of satisfactory min- 
imum strengths. This would result in more costly con- 
crete. Separation of the aggregate into a number of size 
fractions and recombining when batching would increase 

the uniformity of concrete produced. On sizable jobs, 
sufficient savings would result from increased uniformity 
of concrete to more than offset the cost of processing the 
aggregates. More emphasis should be placed on uniform- 
ity of grading of the fine aggregate, since variations in 
grading of the fine aggregate influence workability to a 
greater degree than variations in grading of the coarse 
aggregate. 

Particle shape exerts an influence mainly on the a- 

mount of paste necessary to achieve workability. Flat 
and elongated particles require a greater paste content 
than rounded or cubical particles for the same workabil- 
ity. Significant savings in concrete costs can be effected 
by reducing flat and elongated pieces to a minimum. 



A5-5 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



In the case of aggregate manufactured by crushing, par- 
ticles more nearly cubical aie to be preferred over an- 
gular panicles, the shape being generally dependent on 
the type of crusher being used. 

Different types of aggregates may, at equal water- 
cement ratios, produce concretes of different strength. 
Some of this difference may result indirectly from the 
effect of certain aggregate characteristics such as surface 
texture on the amount of cement paste in the mix. Smooth 
surfaces require less paste for lubrication than rough sur- 
faces. The main differences in strength attributable to 
aggregate type are thought to be due to the degree of 
bond developed between the paste and aggregate (both 
porosity and surface texture being important) and to the 
strength and elastic properties of the aggregate particles 
themselves. The interplay of characteristics such as po- 
rosity, surface texture, strength, elastic properties, and 
other variables makes it difficult to predict behavior. 
Bond of the paste to aggregate particles generally in- 
creases with roughness, although many other aggregate 
characteristics also influence bond. Increase in bond 
due to surface roughness, or for any other reason, will 
increase the flexural strength of the concrete. However, 
differences in surface texture have relatively little in- 
fluence on compressive strength. 



Lightweight Aggregates 

Lightweight aggregate, generally produced by arti- 
ficial means, is assuming increasing importance in pre- 
stressed concrete work, and, for that matter. In most 
phases of concrete construction. These aggregates are 
obtained by expanding, calcining, or sintering materials 
such as blast-furnace slag, shale, slate, clay, diatom ite, 
and others. Differences in the elastic properties of light- 
weight aggregate can cause more change in strength and 
elasticity of the concrete than those of the heavier nat- 
ural aggregates such as gravels and crushed stones. Al- 
though, like normal-weight concrete, water-cement ra- 
tio is the main controlling influence in strength develop- 
ment, an important factor is the strength of the particles 
of lightweight aggregate. 

One of the reasons for the apparent wide differences 
in strengths produced by different lightweight aggregates 
lies in the large and not readily determinable absorption 
characteristics of these materials. For this reason, mixes 
are usually compared on an equal cement -content basis, 
rather than on an equal water -cement ratio basis. This 
may account for a large pan of the strength differences 
attributed to the aggregate, per se. Lightweight -aggre- 
gate concretes differ considerably in the amount of water 
required to attain proper workability, due to grading, 
shape, and surface texture of the aggregate used. Those 



aggregates which, for the reasons noted, require exces- 
sive amounts of water to produce proper workability will 
therefore lequire excessive amounts of cement in order 
to provide adequate strength. The use of natural sand to 
replace a portion of the fines or the use of air entrain - 
ment will result in reduced water requirements. The 
natural sand will raise the unit weight of the concrete 
and decrease the advantages accruing to lower weight. 

Concrete made with lightweight aggregate has gen- 
erally been said to have an upper limit of compressive 
strength of about 5000 psi. Advances in the technology 
of manufacture of such aggregates, closer control of uni- 
formity of grading, and a greater appreciation of pro- 
portioning and mixing techniques have served to raise 
this limit considerably. Early high -strength concrete can 
be produced with most lightweight aggregates in a man- 
ner similar to the heavier natural gravels and crushed 
stones. Some work by our laboratories in this regard will 
appear this year in the Journal of the American Concrete 
Institute (2), 

INFLUENCE OF CONTROL OF UNIFORMITY 

When representative samples of a concrete are sub- 
jected to strength tests, the results of these tests show a 
certain dispersion about an arithmetic mean. This is not 
a peculiarity of concrete alone, but is a common occur- 
rence with other building materials. These differences 
in strength of presumably representative test cylinders 
made from supposedly like batches of concrete may re- 
sult from either random or systematic variation of nu- 
merous factors. A few, for example, are variations in 
batching and mixing, curing of cylinders, sampling of 
concrete, compaction, and capping and testing tech- 
niques. 

The degree to which effective control measures 
are adhered to in minimizing variations determines the 
magnitude of the spread or dispersion of test results about 
the arithmetic mean. In order to assure that more than 
a certain small percentage of the concrete being pro- 
duced does not go below a specified minimum design 
strength, the average strength must of necessity be high- 
er than this minimum. The difference between this av- 
erage strength and the minimum is influenced by the 
degree of control of all phases of concrete making, sam- 
pling, curing, and testing. The less effective the control 
exercised, the higher the average strength required to 
ensure attaining the proper percentage of tests higher than 
the minimum design strength designated. When produc- 
ing early high-strength concrete for prestressing, design- 
ing for a considerably higher strength than required in 
order to offset poor control is particularly uneconomical. 

The level of control required for prestressed con- 
crete work should be higher than that for most other con - 



A5-6 



EARLY HIGH-STRENGTH CONCRETE FOR PRE-STRESSING 



crcte construction. A CI Committee 214^ recommends 
for most concrete construction that the control of con- 
crete-making operations be such that not more than one 
test in ten fall below the designated minimum strength. 
For prestressed concrete work a higher degree of control 
appears desirable, possibly as a goal one test out of one 
hundred. In order to attain this goal as economically as 
possible and not need to over-design the concrete drasti- 
cally to attain the requirement of one in one hundred, 
strict attention must be paid to all phases of the opera- 
tion to reduce variations to a minimum. The lower the 
coefficient of variation of test results, regardless of the 
probability level adopted, the lower the average strength 
required to assure attaining the necessary minimum 
strength of concrete. For example, considering the cri- 
terion of one test out of one hundred falling below a des- 
ignated minimum strength, with a coefficient of varia- 
tion of 10% (considered excellent for general field con- 
struction), the average concrete strength necessary would 
be about 31% higher than this minimum. For a coeffi- 
cient of variation of 5%, the average concrete strength 
necessary would be only 14% higher than the minimum 
strength. The savings in concrete cost would help off- 
set any additional costs incident to closer control of op- 
erations. 



OTHER PROPERTIES OF EARLY HIGH-STRENGTH 

CONCRETE 

In the design of prestressed concrete units, physical 
properties of concrete aside from strength are also of im- 
portance. Properties such as modulus of elasticity, creep, 
drying shrinkage, bond to steel, and durability under 
various conditions of exposure must also be considered. 
Considerable information relative to these properties is 
available for concretes of all types used in varied struc- 
tural applications. Most of this information has been 
obtained for concrete strengths ranging up to 5000 or 
6000 psi. The increasing emphasis on early high -strength 
concrete is pointing up the need for more data concerning 
these properties. Data obtained in some of our recently 
completed laboratory studies and current studies afford 
additional background for the discussion to follow. 

Modulus of Ela sticity 

The modulus of elasticity of concrete depends upon 
the modulus of the cement paste, the modulus of the ag- 
gregate, and the relative amounts of these two compo- 
nents. The modulus of elasticity of the paste component 
increases as the degree of hydration increases. Changes 
in modulus for a given concrete occur because of changes 
in the modulus of elasticity of the paste as curing con- 



tinues. As the modulus of the paste increases, the con- 
crete strength also increases, and for any given concrete 
mixture and curing condition there then exists a general 
empirical relationship between strength and modulus of 
elasticity. For the same cement and aggregate, this em- 
pirical relationship between strength and modulus is in- 
fluenced by the relative amounts of paste and aggre- 
gate^. The moduli of elasticity of pastes range up to 
2. 5 to 3. 5 x 10" psi, while those for aggregates are gen- 
erally considerably higher, except for most of the light- 
weight aggregates. Therefore, the greater the volume 
of paste per unit of aggregate, the lower the modulus of 
elasticity should be at comparable degrees of hydration 
or strength. Data supporting this are shown in Fig. 12. 
Here we see the influence of the volumetric ratio of 
paste to aggregate on the modulus of elasticity for dif- 
ferent levels of compressive strength. This merely em- 
phasizes the fact that strength, by itself, is not a good 
indicator of modulus of elasticity of different concrete 
mixtures over the whole range of strengths possible to 

produce with the same materials used in different pro- 
portions. What is important is how the strength was ob- 
tained for any given set of materials. Was it by longer 
curing, by a change in mix proportions, or by a change 
in water-cement ratio? However, for any particular mix 
strength serves as a good indicator of the modulus of 
elasticity, the modulus increasing with strength. Tht 
modulus does not increase as rapidly as strength and ap- 
pears to approach some limiting value in the high -strength 
region. 

The modulus of elasticity of concretes made with 
lightweight aggregates is not influenced materially by 
the volumetric concentration of aggregate since the mod- 
ulus of the aggregate is generally of the same order of 
magnitude as that of the paste(2). The modulus of e- 
lasticity of concretes made with lightweight aggregate 
is considerably lower than that for concretes made with 
sand and gravel or crushed stone as aggregates. Depend- 
ing upon the aggregate, the percentages may range from 
about 40 to 85% at 28 days, and at 6 months or later 
about 40 to 65%. 



Drying Shrinkage 

In prestressed -concrete construction, drying shrinkage 
should be kept to the minimum possible in order to avoid 
excessive loss of prestress. Shrinkage of concrete is in- 
fluenced by many factors, most important of which are 
the amount of water per unit volume of the concrete, the 
elastic pro pen ies of the aggregate, and paste content and 
characteristics. The size and shape of the member and 

the type of exposure are factors, but changes in these 
factors axe generally not possible. 



A5-7 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



In concretes having cement contents in the range of 
4 to 6sk, per cu. yd. , the amount of water per unit vol- 
ume is essentially constant for a particular level of con- 
sistency. In mixes richer than about 6 sk. per cu, yd. f 
the amount of water required for the same consistency 
and workability increases markedly. Rich mixes are gen- 
erally required for early high-strength concrete. The 
water content of these mixes can be reduced provided 
mechanical means of compaction are available. This 
reduction in water content will result in lower drying 
shrinkage. This is a particularly effective means of 
counteracting the inherently higher shrinkage of light- 
weight-aggregate concretes. 

Using the largest maximum size of aggregate that 
will satisfy placing conditions will result in a lower unit 
water content and will accordingly, for a particular wat- 
er -cement ratio, provide more aggregate per unit vol- 
ume. Aggregate provides considerable restraint to shrink- 
age of the paste. The more aggregate per unit volume 
the more restraint to shrinkage. The elastic properties 
of the aggregate play an important role in drying shrink- 
age. Those aggregates having a low modulus of elastic- 
ity provide less restraint than those of higher modulus; 
witness the relatively large drying shrinkage of most 
lightweight -aggregate concretes. Selection of the larg- 
est possible maximum size of aggregate having a rela- 
tively high modulus of elasticity will result in reduction 
in drying shrinkage. 

High-strength zero-slump concretes generally will 
show somewhat lower drying shrinkage than the usual 
structural -grade concrete of plastic consistency. In rec- 
ent tests(2), concretes made with a natural sand and 
gravel at a consistency of about 1\ in. slump (2000 to 
3500 psi at 7 days) showed an average drying shrinkage 
at one year of about 680 millionths. For the same ag- 
gregate in zero -slump concretes compacted by vibra- 
tion (6000 to 8000 psi at 7 days) the average drying 
shrinkage was about 550 millionths. For these two class- 
es of concrete, but made with a typical expanded -shale 
ligl eight aggregate, the respective drying shrinkage 
averages were 900 and 600 millionths. 



reep 

The < reep of concrete under sustained load results in 
a loss in prepress, as does the shortening due to drying 
shrinkage. Considering this point alone, creep should 
be kept to a minimum. Creep does serve a useful func 
Uon, howe , in relieving localized overloads by redis- 
tribution and equalization of stress. 

Most available data indicate that creep of concrete 
laveitely proportional to its strength. Creep has been 



related to many other variables such as age, curing, and 
type of cement. These may be indirect relationships. 
The physical characteristics of the aggregate are of im- 
portance and operate in a manner similar to their effect 
on drying shrinkage. The moisture content of the con- 
crete is an important factor; concrete which is drying 
exhibits greater creep than wet concrete. 

Many of these means of reducing creep cannot al- 
ways be utilized. From a practical standpoint, using 
certain available materials for producing concrete, what 
are the creep characteristics of high -strength concrete 
made from these materials? Recent tests^ 2 ^ of concretes 
made with natural sand and gravel and with an expanded- 
shale lightweight aggregate show that the ultimate creep 
coefficient (millionths per psi) as a function of com- 
pressive strength was not linear over the range of strength! 
from 2000 to 10,000 psi. The relationship departed 
from linearity at about 5000 to 6000 psi, the higher- 
strength concretes showing less of a reduction in ultimate 
creep coefficient with increase in strength than those 
below 6000 psi. For 7-day loading, at 8000 psi, the ul- 
timate creep coefficient for the sand and gravel con- 
crete was about 0.40 millionths per psi, at 6000 psi it 
was about 0.55, at 4000 psi 0.80, and at 2000 psi 1.40 
millionths per psi. The values for the expanded -shale 
lightweight -aggregate concretes were about the same at 
strengths below 5000 to 6000 psi. At the higher strength 
levels, these concretes showed greater ultimate creep 
coefficients than the sand and gravel concrete. Most 
other lightweight aggregates included in this study showed 
greater creep coefficients at all strength levels than the 
sand and gravel concrete, while a few showed slightly 
less creep. 

Bond of Concrete to Steel 

Bond of concrete to steel increases with increase in 
compressive strength of the concrete. The relationship 
is curvilinear, bond strength increasing less rapidly as 
the compressive strength of the concrete is increased. 
The bond strengths developed by high-strength concretes 
are far above the maximum allowable average bond 
stress recommended by ACI-ASCE Joint Com mittee 
*23 ( I. in some recent tests^ 2 ) of vibrated zero -slump 
concretes in the strength range of 7000 to 10,000 psi, 
bond strengths of top horizontal bari> in pull-out speci- 
mens exceeded 1400 psi. Actually, the yield strength 
uf the steel was developed in most of these tests, indi- 
cating higher bond strengths than this figure. This was 
for the sand and gravel concretes and one of the light- 
weight-aggregate concretes (expanded shale). At the 
3000 to 4500 -psi level, bond strengths developed for top 
horizontal bars ranged from 520 to 900 psi. Included in 



A< 



EARLY HIGH-STRENGTH CONCRETE FOR PRESTRESSING 



this Latter strength range were seven different light- 
weight-aggregate concretes and one sand and gravel con- 
crete. 

Durability 

In many applications, prestressed concrete will be 
subjected to severe exposure conditions. Bridge decks, 
for example, will be exposed to the use of salts for pur- 
poses of ice removal. Portions of structures may be ex- 
posed to freezing and thawing while in a saturated state. 
These severe service conditions must be met by provid- 
ing concrete with a high degree of resistance to such ex- 
posure. 

Water in concrete, like any other water, undergoes 
about a 9% increase in volume when frozen. This freez- 
ing in concrete is a gradual process as the temperature 
drops because of rate of he at transfer, difference in freez- 
ing point of water in the pores of various sizes, and the 
progressive increase in concentration of dissolved alkalies 
in the remaining unfrozen paste liquid. Freezing creates 
an excess volume of water which must reach some relief 
zone, such as an unsaturated region or an air void, or 
disruptive hydraulic pressures will be created. Factors 
which determine the potential destructiveness of this 
freezing are the amount of water freezing, the permea- 
bility of the paste, the tensile strength of the concrete, 
the rate of freezing, characteristics of the aggregate, 
and -- most important of all --the amount and distribu- 
tion of air voids in the paste. 

Based on extensive laboratory data and field service 
records, we know that the only practical and efficient 
means of securing a high level of resistance to freezing 
and thawing is by entraining air in the required amount 
and proper distribution of air void sizes. In mixes of 
plastic consistency, the proper amount of air is based on 
a requirement of 9± 1% of air in the mortar fraction of 
the iniXV*', For concretes made with aggregate of max- 
imum size l{ in. , the mortar content is approximately 
50 to 55rfo by volume. On a concrete basis, the air con- 
tent would be4^ to 5%. For other maximum sizes of ag- 
gregate, the required air content would be calculated in 
the same manner. Control of air is then based on the 
determinations of the air content of the concrete as a 
whole. 

For the same air -entraining cement or the same a- 
mount of air -entraining admixture per unit of cement, 
decreasing the slump within the range of slumps normal- 
ly used results in decreased volumes of air entrained. 
Some laboratory tests made using both a tilting -drum 
mixer and an open -tub type mixer and a typical Type LA 
Portland cement showed air contents of about 4^Jo at a 
slump of 3 in. , while at a slump of 4 to £ in. the air 



content was slightly over 3% Internal vibration of rela- 
tively long duration lowered the air content to about 3fy 
and 27o. respectively. While these are considerable re- 
ductions on a volume basis, the resistance to freezing 
and thawing is primarily dependent on the size and dis- 
tribution of the air voids, not on the gross amount. Vi- 
bration removes the large voids, which account for a 
greater proportion of the volume of the air, and leaves 
the average size and spacing of the small, but efficient, 
voids relatively unchanged. 

Curing air -entrained concretes at elevated temper- 
atures does not appear to reduce the beneficial influ- 
ence of the entrained air on durability. Recent labora- 
tory tests comparing air -entrained concretes made and 
cured at 70°F and at 105°F indicated comparable resis- 
tance to freezing and thawing, even though at the ele- 
vated temperature of mixing the volume of air entrained 
was reduced slightly. In fact, for those concretes whose 

curing period prior to test included an air-drying period, 
the 105°F air-entrained concretes were somewhat more 
durable than the 73°F concretes. 

Entrained air increases the durability of concretes 
made with any of the types of portland cements or port- 
land blast-furnace slag cements. In some recent tests of 
resistance to scaling(9) resulting from the use of salts for 
ice -removal purposes, ihe use of entrained air in con- 
cretes made with either Type I, 11, or [II portland ce- 
ment produced concretes of comparable and greatly im- 
proved resistance to scaling. These particular air -en- 
trained concretes were made by adding an air -entraining 
admixture at the mixer. Similar results would have been 
obtained by the use of the respective a 11 -entraining ce- 
ment types. 

Aggregates of good quality are a necessary compo- 
nent of durable concrete. Unsound aggregate particles 
such as chert, soft sandstone, clay lumps, and some high- 
ly absorptive limestones may, when saturated and incor- 
porated in even an air-entrained paste, cause disruptic 
of the concrete during freezing. Some of these aggre - 
gates may be beneficiated by drying prior to use or by 
removal of the offending particles by methods such as 
heavy-media separation or an elastic -rebound processing 

technique. 



SUMMARY 

Early high-strength concrete can be produced by any 
one of a number of techniques or combinations of tech- 
niques. The selection of method will depend upon the 
availability of specific materials, mixing equipmer-r, 
and compacting equipment The following means for 
obtaining concrete of high quality and having early high 



A5-9 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



strength for prestressing are suggested, together with oth- 
er considerations; 



1. 



2. 



3. 



4. 



5. 



6. 



The use of low water -cement ratio, rich mixes of 
plastic consistency, or the use of low water-cement 
ratio, rich, dry mixes placed by mechanical means 
of compaction such as vibration, to permit the use 
of more aggregate per unit volume. 

The use of high -early -strength port land cement 
(Type III). 

The use of saturated steam at atmospheric pressure 
at temperatures below the boiling point of water, 
together with the use of effective insulation to re- 
tain heat generated by cement hydration. 

Careful control of aggregate gradation, weighing of 
materials, and mixing, compacting, and curing of 
concrete. 



The use of entrained air to assure adequate resis- 
tance to exposures involving freezing and thawing 
and the use of de-icing chemicals. 

The use of water curing during the early hours of 

hydration, either by ponding the surface or by the 
use of saturated aggregates. When dry or saturated 
lightweight aggregates are used, the provision of 

additional curing water does not appear necessary 

since aggregates of this type hold greater amounts 

of water available for this purpose. 



the Journal of the American Concrete Institute. Pre- 
sented at the ACI Convention in Dallas, Feb. , 1957. 

4. Verbeck, George J. , and Foster, Cecil W. : "The 
Long-Time Study of Cement Performance with Special 
Reference to Heats of Hydration, " Proc, ASTM, Vol. 
50, 1950. Reprinted as PCA Bulletin 32. 

5. "Recommended Practice for Evaluation of Compres- 
sion Test Results of Field Concrete (ACI 214-57), " Jour- 
nal of the American Concrete Institute, Proc, Vol. 54, 
July, 1957. 

6. ACI-ASCE Joint Committee 323 , "Recommended 
Practice for Prestressed Concrete, " December, 1956. 

7. Klieger, Paul: "The Long -Time Study of Cement 
Performance in Concrete, Chapter 10 - Progress Report 
on Strength and Elastic Properties of Concrete, " to be 
published in the Journal of the American Concrete in- 
stitute. 

8. Klieger, Paul: "Effect of Entrained Air on Strength 
and Durability of Concretes Made With Various Maxi- 
mum Sizes of Aggregate, "Highway Research Board Pro- 
ceedings, December 1951. Reprinted as PCA Bulletin 40. 

9. Klieger, Paul: "Curing Requirements for Scale Re- 
sistance of Concrete, " Highway Research Board Bulletin 
128, 1956. Reprinted as PCA Bulletin 82. 

SELECTED BIBLIOGRAPHY 



7. Although calcium chloride will increase the early 
strength development of concrete, in view of the 
possibility of serious corrosion of the prestressmg 
steel its use in prestressed concrete is not now rec 
om mended. 



REFERENCES 

1. Powers, T.C. : "A Discussion of Cement Hydration 
in Relation to the Curing of Concrete," Highway Re- 
search Board Proceedings Vol. 27. 1947. Reprinted as 
PCA Bulletin 25. 

2. Shideler, Joseph J. • "Lightweight Aggregate Con- 
crete for Structural Applications, "to be published in the 
October 1957 Journal of the American Concrete Institute 
Presented at the ACI Convention in Dallas, Feb. , 1957. 

3. Klieger, Paul: The Effect of Mixing and Curing 
Temperature on Concrete Strength, " to be published in 



Banks, R, R. • "High Strength Concrete Required for Pre- 
stressing, " Commonwealth Engr. 41, (2) 48-51 (Sept. 
1953); Abstract in Eng. Ind. (1953). 

Banks, K. R. : "High Strength Concrete for Prestressing, ■ 
Constructional Review^, (3) 17-25 (1953); Abstract in 
B. S. A. 26 (9) 1145 (Sept. 1953). 

Banks, K. R. ■ "High Strength Concrete for Prestressing, " 
Symposium on Prestressed Concrete, Sydney, June 30 - 
July 2, 1953, Cement and Concrete At§0< of Australia, 
pages 40-47; discussion, pages 48-54. 

Blanks, R. F. : "Concrete for Prestressing," First U.S. 
Conferew e on Prestressed Concrete (see B.S.A. 1538) 
136-49 (1951); Abstract in B. S.A. 24 (12) 1671 (Dec 
1951), 

Carlson, R. W. : "Drying Shrinkage of Concrete as Af- 
fected by Many Factors, M Proc ASTM, Vol. 38, Pan 
II. 1938. 



A5-I0 



EARLY HIGH-STRENGTH CONCRETE FOR PRESTRESSING 



Collins, A.R. : "The Principles of Making High -Strength 
Concrete," Engineering 169 (4382) 80-1 (Jan. 20, 1950). 



Davis, R. E. and Troxell, G. E, : "Properties of Concrete 
and Their Influence on Prestress Design, " Journal of the 
American Concrete Institute, January, 1954, (Proc. Vol. 
50). 



Kennedy, H. L. : "High Strength Concrete," First U.S. 
Conference on Prestressed Concrete (see B. S. A, 1538) 
126-35 (1951); Abstract in B.S.A, 24 (12) 1670 (Dec. 
1951). 

Menzel, Carl A.: "An Investigation of Bond, Anchor- 
age, and Related Factors in Reinforced Concrete Beams, " 
Portland Cement Association, Research Department Bul- 
letin No. 42. 



> 






A5-I1 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



e> 

CO 

UJ 



en 
</> 

UJ 
Gl 



o 



VIBRATED 



HAND- PLACED 



FULL COMPACTION 



WATER -CEMENT RATIO 



Fig. 1 Influence of W/C on Strength 



NET W/C, GAL/SK 





3.5 4.0 4.5 




! 1 1 1 V 1 r 


d l3 

>- 


EXTERNAL VIBRATION - 
4900 VPM 


D 12 


- ©(20% SAND, BY WT.) 


^ 






\ 


i 


V* 




o 



2 ,0 
UJ 

9 



8 



UJ 7 

5 

UJ 

O 6 



tie 




(37%) 



i 



1 



1 



1 



1 



i 



(27%) 



028 030 032 34 0.36 38 040 42 

NET W/C - by wt. 

Fig. 2 Influence of Method of 
Compaction on Cement Requirements 



9 

o 
o 

2 e 



fi 7 - -~- 






£ 



en 4 

UJ 

07 5 

CO 

UJ 

£ a 

o 







NET W/C, GAL/SK 

3 5 4.0 4.5 

1 



T 



T 



T 



r 



PONDED 




_^-> BURLAP 



280 



PONDED 



TYPE I CEMENT 




BURLAP 



— DRY AGGREGATE 

SATURATED AGGREGATE 

i 1 * 



ID 



028 030 032 34 036 038 040 042 

NET W/C, BY WT 

Fig. 3 Influence of Moisture 
Availability on Strength Development 
(Hand-Placed Concretes) 



o 

° 9 

o 9 



C 8 



7 



(D 



UJ 4 
> 

CO 

(/) 3 
UJ 

or 

CL 9 



o 







NET W/C t GAL/SK 

3 5 4.0 4 5 

1 1 



T 



T 



T 



PONDED 




BURLAP 



PONDED 



BURLAP 



28 D 




3D 



I D 



AGGREGATE SATURATED 



1 



TYPE I CEMENT 



i 



i 



i. 



i. 



028 030 032 034 Q36 0.38 040 Q42 

NET W/C, BY WT 

Fig, 4 Influence of Moisture 
Availability on Strength Development 
(Vibrated Concretes) 



A5-I2 



EARLY HIGH-STRENGTH CONCRETE FOR PRE-STRESSING 



T 



T 



T 



180 

\L 

• 160 

140 



I 120 

O 

Z 100k 

X 

h 80h 



& 



60 



S 40 
U 

2j 20 








ONE* DAY TESTS 
TYPE I CEMENT 



I 



30 40 50 60 70 80 90 100 IK) 120 

MIXING a CURING TEMPERATURE -°F 
Fig. 5 Influence of Mixing and Curing Tem- 
perature on Early Strength Development 




2 4 



8 10 12 14 16 
AGE IN HOURS 



8 20 22 24 



8 



o 

o 

o 



IP 

I 
I 



INSULATION 

I" THICK SPUN 
GLASS BLANKET 



y 



UJ 

E 



> 

UJ 



O 



TYPE m 
CEMENT 




[INSULATE 
Jl DAY THEN 
(MOIST AT- 
\73 # F 



Fig. 6 Influence of Heat Generated Dur- 
ing Hydration on Concrete Temperature 
(Calculated Assuming Full Retention of 
Heat Generated) 



TYPE I 
CEMENT 



INSULATE Q 
OAT THEN 
MOIST AT 
Z3 # F 



r 



a 

8 8 



en 7 
a. 



i 



6 







NET W/C - 36 
TYPE 1-64 SK/CU YD 
TYPE HI- 6 9 SK/CU YO 
DRY, VIBRATED 
6 * 12 -IN CYLINDERS 






4 



UJ 3 
co 2 

UJ 



o 



I 







r&2— 



(1.23) 




73° F a 100% RH . 



NET W/C - 35 
ZERO SLUMP 
VIBRATED 



i. 



i 



x 







AGE - DAYS 



2 3 4 5 6 

AGE - DAYS 



Fig. 7 Influence of Heat Insulation on 
Strength Development 



Fig. 8 Influence of Cement Type on 
Strength 



A5-I3 



WORLD CONFERENCE ON PRESTRESSED CONCRETE 



o 
o 
o 8 



m 7 

I 

I 6 

h- 

O 

Z 5 

UJ 

4 



^CJ *— 



< Lie) - 



r, 



"(146) 



*° 



G<* 



&z_ 



CO 

UJ 

> 3 h 

CO 

co 

UJ 

cc 

CL 



. (196} 



o 
o 



I - 







u 



73° F 6 100% RH. 



TYPE I CEMENT 
NET W/C- 35 

ZERO SLUMP 

VIBRATED 



1 



1 



1 



i 



X 



1 



1 



Fig. 
Deve 



l 234567 

AGE - DAYS 

9 Influence of CaCl^ on Strength 
lopment 



o 
o 
o 



to 

CL 



I 



7 - 



6 - 



5 - 



4 - 



3 - 



O 2 - 



UJ 

cr 

CO 
UJ 



I - 









^ 2 - 



NET W/C, GAL/SK 
3.5 40 45 



i 
O 

z 

UJ 



T 



CO 3 

UJ 

> 

CO 2 
CO 

LU 



£ 




O 








, . TYPE I CEMENT 

• I -2 INCH SLUMP. HAND PLACED 

• SLUMP, VIBRATED 



1 



X 



1 



1 



028 030 0.32 0.34 036 038 040 042 

NET W/C , BY WT. 
Fig. 10 Hand-Placed vs. Vibration 



O 

V)-' 



8 - 



7 - 



OUJ 



T 



T 



T 



T 




10,000 

9,000 

8,000 



" 



* 7.000 



2 02O 



j. 



TYPE I CEMENT „ 

NATURAL SAND AND 

_, GRA,VEL 



030 



40 



050 



RATIO OF PASTE TO AGGREGATE -By vol 



4900 7200 II 000 FREQ.vPtt 

E E F F G 6 AMP 

E '- LARGE F -- MEDIUM G = SMALL 

Fig. 11 Influence of Frequency 8, Fig. 12 Influence of Compressive 

Amplitude of Vibration on Strength Strength and Mix Proportions on 

(External Vibration) Modulus of Elasticity 



060 



A5-J4 



>